get_transcript
ChatGPTRead a finished transcript. format "text" gives one line per segment with optional [m:ss] timestamps and speaker labels; "srt" or "vtt" give subtitle files. Very long transcripts are cut off, with a link to the full one.
Overview
ScribeToAny turns audio and video into text. Connect your ScribeToAny account to transcribe files and links from ChatGPT and to read the transcripts you already have. What you can do: - Attach an audio or video file in the chat and ask ChatGPT to transcribe it. - Transcribe a link from X, Twitch, Loom, Google Drive or Dropbox, or a direct link to an audio or video file. - List and search your transcripts, check whether a transcription has finished, and get the full text as plain text with or without timestamps, or as SRT or VTT subtitles. - See how much of your plan's transcription allowance is left. Transcription runs on ScribeToAny with speech recognition in about 98 languages and speaker labels. Each transcript also stays in your ScribeToAny dashboard, where you can edit, export and share it. You need a ScribeToAny account. Transcriptions started from ChatGPT count toward your plan like any other. The app can read your transcripts, their titles and file names, and your usage, and it can start new transcriptions; it can't edit or delete anything.
Tools
get_transcription_status
ChatGPTCheck whether a transcription has finished. Status is queued, transcribing, done or failed (with the error). Long recordings can take a few minutes.
get_upload_link
ChatGPTGet a ScribeToAny page where the user can upload an audio or video file from their device to transcribe it. Use this when the user wants to transcribe a local file and can't attach it here. The user must be signed in to the same ScribeToAny account in the browser. Afterwards the new transcript is first in list_transcripts; poll get_transcription_status with its id.
get_usage
ChatGPTShow the user's plan and how much of their transcription allowance is left: files transcribed today and minutes used this billing month.
list_transcripts
ChatGPTList the user's ScribeToAny transcripts, newest first, with id, name, status (queued, transcribing, done, failed), language and duration. Use the id with get_transcript.
transcribe_file
ChatGPTStart transcribing an audio or video file the user attached in this ChatGPT conversation (MP3, WAV, M4A, MP4, MOV, WEBM…). Only for files shared in the chat; for a public link use transcribe_link, and in apps that can't attach files use get_upload_link. Returns a transcript id; poll get_transcription_status until it is done. Uses the account's transcription allowance.
transcribe_link
ChatGPTStart transcribing audio or video from a public link: X (Twitter), Twitch, Loom, Google Drive or Dropbox share links, or a direct link to a media file. YouTube, TikTok and Instagram links are not supported (download the file and upload it on scribetoany.com instead). Returns a transcript id; poll get_transcription_status until it is done. Uses the account's transcription allowance.
Capabilities
Writes
Example Prompts
Click any prompt to copy it.
